An econometric analysis for the energy efficiency indicators in thermal power plants based upon the GARCH model

A new hybrid methodology is introduced in this paper for analysing the energy efficiency indicators of thermal power plants. The multiple regression model and the generalised autoregressive conditional heteroskedasticity (GARCH) model are composed in this model. There are five indicators variables are employed. The examination results indicate that this new hybrid econometric analysis can not only estimate the energy efficiency indicators without manual intervention, but analyse the short-term volatility of variables. Moreover, the methodology can eliminate the disturbance of residual superposition. Furthermore, the method can provide direct information on the degree of volatility among determinants and energy efficiency indicators. The accuracy of hybrid methodology has been proved by the comparison results of vector auto-regression (VAR) model.